Re: Universal grammar

In article <1161214693.444791.304000@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>, "Peter
T. Daniels" <grammatim@xxxxxxxxxxx> wrote:

I found:


http://www.cogsci.umn.edu/OLD/calendar/past_events/millennium/files/1030123953.html
It says:
Whereas Chomsky speaks of broad universals of human language,
Greenberg et al.
attempt to find structural and grammatical universals in a wide
range of human
languages.
The "Chomsky hierarchy" shows up as a classification of formal

That comes from the 1950s, had little or no influence on linguistics
and the study of human language, but apparently is important to AIists.

mathematical grammars. The more general formal grammars are though
in practice not used so much in favor of so called context free grammars
(CFG) combined with context dependencies implemented using semantic
actions attached to the grammar rules.

Using this language, I am more at Chomsky moving towards Greenberg rather
than the opposite direction. It is just a personal choice and adaption.

So why didn't you say you're not interested in human language?

I am not sure what you are saying here. As far as I know, the same formal
grammar theory is used in the parsing of human and computer languages.
In practice hybrid methods are used in favor as opposed a clean
grammar theory expressing context dependancies, for example. So the more
general Chomsky type grammars haven't been used much in computer parsing
either. There is no difference here between human and computer languages
here, but in the complexity.

BTW "word etymology" is not the province of linguists. It's done by
philologists and lexicographers.

Thank for the clarification - I have a hazy understanding of such things.
:-) The reason I mentioned it though, was that some of the Greenberg
references I gave earlier in this thread indicated that he used etymology
for his some of his conclusions.

Of course! How else would you do comparative linguistics?

Linguists don't _compose_ etymologies, but they rely on existing
dictionaries, Greenberg a lot more credulously than most. There is a
_huge_ literature on how wrong he went beginning ca. 1987.

So I might make use of some such work, as an input, but I am unlikely do
do any research in it.

--
Hans Aberg
.

Relevant Pages

  • Re: Universal grammar
    ... The "Chomsky hierarchy" shows up as a classification of formal ... and the study of human language, but apparently is important to AIists. ... Computer languages have much more restricted grammars than human languages, ... But the 'complexity' is a qualitative difference, ...
    (sci.lang)
  • Re: Who castrated Esperanto?
    ... languages. ... exact nature of UG, but that it exists is not in dispute. ... It is NOT a fact that what Chomsky labeled "UG" exists. ... how to speak and understand human language. ...
    (sci.lang)
  • Re: word 2007 and grammar check
    ... you that fluent computer use of human language is decades, ... It has several dialects where words have totally different meanings depending upon usage from one dialect to another and from one geographic region to another. ... The best that you can do is to have proofing tools appropriate to the work that you are doing Even then, grammar checking will cause headaches. ...
    (microsoft.public.word.docmanagement)
  • Re: Natural language programming?
    ... Human language can express ... (not to mention they aren't context free in the grammar sense). ... > of how the car works technically. ... then he shouldn't be programming in it. ...
    (comp.programming)
Loading